{"id":"https://openalex.org/W4410584357","doi":"https://doi.org/10.23919/date64628.2025.10993131","title":"BMP-SD: Marrying Binary and Mixed-Precision Quantization for Efficient Stable Diffusion Inference","display_name":"BMP-SD: Marrying Binary and Mixed-Precision Quantization for Efficient Stable Diffusion Inference","publication_year":2025,"publication_date":"2025-03-31","ids":{"openalex":"https://openalex.org/W4410584357","doi":"https://doi.org/10.23919/date64628.2025.10993131"},"language":"en","primary_location":{"id":"doi:10.23919/date64628.2025.10993131","is_oa":false,"landing_page_url":"https://doi.org/10.23919/date64628.2025.10993131","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 Design, Automation & Test in Europe Conference (DATE)","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5091041998","display_name":"Cheng Gu","orcid":"https://orcid.org/0000-0001-5259-6203"},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Cheng Gu","raw_affiliation_strings":["Shanghai Jiao Tong University,Department of Computer Science and Engineering,Shanghai,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Shanghai Jiao Tong University,Department of Computer Science and Engineering,Shanghai,China","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100438696","display_name":"Gang Li","orcid":"https://orcid.org/0000-0002-4549-293X"},"institutions":[{"id":"https://openalex.org/I4210112150","display_name":"Institute of Automation","ror":"https://ror.org/022c3hy66","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210112150"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Gang Li","raw_affiliation_strings":["CASIA,The Key Laboratory of Cognition and Decision Intelligence for Complex Systems,Beijing,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"CASIA,The Key Laboratory of Cognition and Decision Intelligence for Complex Systems,Beijing,China","institution_ids":["https://openalex.org/I4210112150"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101112881","display_name":"Xiaolong Lin","orcid":null},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xiaolong Lin","raw_affiliation_strings":["Shanghai Jiao Tong University,Department of Computer Science and Engineering,Shanghai,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Shanghai Jiao Tong University,Department of Computer Science and Engineering,Shanghai,China","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5062205543","display_name":"Jiayao Ling","orcid":"https://orcid.org/0000-0003-2121-2996"},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jiayao Ling","raw_affiliation_strings":["Shanghai Jiao Tong University,Department of Computer Science and Engineering,Shanghai,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Shanghai Jiao Tong University,Department of Computer Science and Engineering,Shanghai,China","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5112149353","display_name":"Jian Cheng","orcid":"https://orcid.org/0000-0003-1289-2758"},"institutions":[{"id":"https://openalex.org/I4210112150","display_name":"Institute of Automation","ror":"https://ror.org/022c3hy66","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210112150"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jian Cheng","raw_affiliation_strings":["CASIA,The Key Laboratory of Cognition and Decision Intelligence for Complex Systems,Beijing,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"CASIA,The Key Laboratory of Cognition and Decision Intelligence for Complex Systems,Beijing,China","institution_ids":["https://openalex.org/I4210112150"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5056632010","display_name":"Xiaoyao Liang","orcid":"https://orcid.org/0000-0002-2790-5884"},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xiaoyao Liang","raw_affiliation_strings":["Shanghai Jiao Tong University,Department of Computer Science and Engineering,Shanghai,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Shanghai Jiao Tong University,Department of Computer Science and Engineering,Shanghai,China","institution_ids":["https://openalex.org/I183067930"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"7"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10862","display_name":"AI in cancer detection","score":0.8942999839782715,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10862","display_name":"AI in cancer detection","score":0.8942999839782715,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.85589998960495,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10901","display_name":"Advanced Data Compression Techniques","score":0.8420000076293945,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.6599941253662109},{"id":"https://openalex.org/keywords/binary-number","display_name":"Binary number","score":0.6525850296020508},{"id":"https://openalex.org/keywords/quantization","display_name":"Quantization (signal processing)","score":0.6298408508300781},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.5587968230247498},{"id":"https://openalex.org/keywords/diffusion","display_name":"Diffusion","score":0.4749089479446411},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.45852792263031006},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3451871871948242},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.2542899250984192},{"id":"https://openalex.org/keywords/arithmetic","display_name":"Arithmetic","score":0.14538851380348206},{"id":"https://openalex.org/keywords/thermodynamics","display_name":"Thermodynamics","score":0.11699086427688599},{"id":"https://openalex.org/keywords/physics","display_name":"Physics","score":0.11359164118766785}],"concepts":[{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.6599941253662109},{"id":"https://openalex.org/C48372109","wikidata":"https://www.wikidata.org/wiki/Q3913","display_name":"Binary number","level":2,"score":0.6525850296020508},{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.6298408508300781},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5587968230247498},{"id":"https://openalex.org/C69357855","wikidata":"https://www.wikidata.org/wiki/Q163214","display_name":"Diffusion","level":2,"score":0.4749089479446411},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.45852792263031006},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3451871871948242},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.2542899250984192},{"id":"https://openalex.org/C94375191","wikidata":"https://www.wikidata.org/wiki/Q11205","display_name":"Arithmetic","level":1,"score":0.14538851380348206},{"id":"https://openalex.org/C97355855","wikidata":"https://www.wikidata.org/wiki/Q11473","display_name":"Thermodynamics","level":1,"score":0.11699086427688599},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.11359164118766785}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.23919/date64628.2025.10993131","is_oa":false,"landing_page_url":"https://doi.org/10.23919/date64628.2025.10993131","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 Design, Automation & Test in Europe Conference (DATE)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/5","display_name":"Gender equality","score":0.46000000834465027}],"awards":[{"id":"https://openalex.org/G1821119093","display_name":null,"funder_award_id":"Z231100010323002","funder_id":"https://openalex.org/F4320325902","funder_display_name":"Beijing Municipal Science and Technology Commission"},{"id":"https://openalex.org/G4679563804","display_name":null,"funder_award_id":"2022ZD0160304","funder_id":"https://openalex.org/F4320335777","funder_display_name":"National Key Research and Development Program of China"}],"funders":[{"id":"https://openalex.org/F4320325902","display_name":"Beijing Municipal Science and Technology Commission","ror":null},{"id":"https://openalex.org/F4320335777","display_name":"National Key Research and Development Program of China","ror":null}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":24,"referenced_works":["https://openalex.org/W1861492603","https://openalex.org/W1901129140","https://openalex.org/W2541839172","https://openalex.org/W2892341857","https://openalex.org/W2949870694","https://openalex.org/W2962834855","https://openalex.org/W3155072588","https://openalex.org/W4250482878","https://openalex.org/W4312497550","https://openalex.org/W4312933868","https://openalex.org/W4360884927","https://openalex.org/W4386057725","https://openalex.org/W4386765191","https://openalex.org/W4390874074","https://openalex.org/W4403791519","https://openalex.org/W4405003117","https://openalex.org/W6640963894","https://openalex.org/W6739901393","https://openalex.org/W6773820404","https://openalex.org/W6791353385","https://openalex.org/W6856696905","https://openalex.org/W6861614106","https://openalex.org/W6862355752","https://openalex.org/W6869992887"],"related_works":["https://openalex.org/W2051487156","https://openalex.org/W2073681303","https://openalex.org/W4409439182","https://openalex.org/W2053286651","https://openalex.org/W2055243143","https://openalex.org/W2181743346","https://openalex.org/W2187401768","https://openalex.org/W2181413294","https://openalex.org/W2605915117","https://openalex.org/W2953338458"],"abstract_inverted_index":{"Stable":[0],"Diffusion":[1],"(SD)":[2],"is":[3,32],"an":[4],"emerging":[5],"deep":[6],"neural":[7],"network":[8],"(DNN)":[9],"model":[10,31,137],"that":[11,98,118,146],"has":[12],"demonstrated":[13],"impressive":[14],"capabilities":[15],"in":[16,35,56],"generative":[17],"tasks":[18],"such":[19],"as":[20],"text-to-image":[21,114],"generation.":[22],"However,":[23],"the":[24,29,52,78,96,113,134],"iterative":[25,85],"denoising":[26,101],"stage":[27],"of":[28,54,84],"SD":[30,68],"extremely":[33],"expensive":[34],"both":[36],"computations":[37],"and":[38,46,81,155],"memory":[39,82],"accesses,":[40],"making":[41],"it":[42],"challenging":[43],"for":[44,66,106],"fast":[45],"energy-efficient":[47],"edge":[48],"deployment.":[49],"To":[50],"alleviate":[51],"overhead":[53],"denoising,":[55,86],"this":[57],"paper":[58],"we":[59],"propose":[60],"BMP-SD,":[61],"a":[62,107],"post-training":[63],"quantization":[64,74],"framework":[65],"hardware-efficient":[67],"inference.":[69],"BMP-SD":[70,119,135],"employs":[71],"binary":[72],"weight":[73],"to":[75,152,160],"significantly":[76],"reduce":[77],"computational":[79],"complexity":[80],"footprint":[83],"along":[87],"with":[88,123],"dynamic":[89],"step-aware":[90],"mixed-precision":[91,121],"activation":[92],"quantization,":[93],"based":[94],"on":[95,112,127,138],"observation":[97],"not":[99],"all":[100],"steps":[102],"are":[103],"equally":[104],"important":[105],"specific":[108],"input":[109],"prompt.":[110],"Experiments":[111],"generation":[115],"task":[116],"show":[117],"achieves":[120],"(W1.73A4.87)":[122],"minimal":[124],"accuracy":[125],"loss":[126],"MS-COCO":[128],"2014":[129],"dataset.":[130],"We":[131],"also":[132],"evaluate":[133],"quantized":[136],"three":[139],"state-of-the-art":[140],"bit-flexible":[141],"DNN":[142],"accelerators,":[143],"results":[144],"reveal":[145],"our":[147],"method":[148],"can":[149],"deliver":[150],"up":[151],"5.14\u00d7":[153],"performance":[154],"3.85\u00d7energy":[156],"efficiency":[157],"improvements":[158],"compared":[159],"W8A8":[161],"quantization.":[162]},"counts_by_year":[],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}